In this paper we introduce a new pruning mechanism, called Similarity Pruning for Probabilistic Opponent-Model (PrOM) Search. It is based on imposing a bound on the differences between two or more evaluation functions. Assuming such a bound exists, we are able to prove two theoretical properties, viz., the bound-conservation property and the bounded-gain property. Using these properties we develop a Similarity-Pruning algorithm. Subsequently we conduct a series of experiments on random game trees to measure the efficiency of the new algorithm. The results show that Similarity Pruning increases the efficiency of PrOM search considerably.